Abstract Most tissues in our bodies and those of other animals are epithelia. Proper tissue architecture and integrity require apical-basal polarity establishment and maintenance, cell-cell and cell-matrix adhesion, and linkage of adhesions to the actin cytoskeleton. These also mediate force generation, allowing cells to change shape and move. Most cancers are of epithelial origin, and in many different cancers mutation or altered expression of polarity and junctional proteins leads to changes in cell polarity, promoting cell migration and cell invasion. Linkage of cadherin-based cell-cell adherens junction (AJs) with the actin cytoskeleton regulate dynamic cell behaviors during development and in cancer. A critical player integrating tissue adhesion and polarity in both mammals and Drosophila is Afadin/Canoe (Cno). Cno is a multidomain protein that in Drosophila plays key roles in processes ranging from apical-basal polarity establishment to maintaining AJ-cytoskeleton linkage during mesoderm invagination, germ-band extension, and collective cell migration during dorsal closure. Here I address two outstanding questions in the field: by what mechanisms does Cno work to control tissue architecture and dynamic cell behaviors, and how do upstream inputs regulate Cno's function? I do so via two Aims: 1) Define mechanisms underlying Cno's ability to link the AJs with the actin cytoskeleton during morphogenesis, and 2) Define how Dizzy, a Rap1 activity regulator, coordinates Cno localization and function during embryonic development, and determine where active Rap1 localizes. My hypothesis in Aim 1 is that Cno's PDZ and F-actin binding (FAB) domains play key roles in Cno's localization and function at many stages but may not be essential for all roles. Using CRISPR/Cas9, I engineered cno's locus to reintroduce a series of mutants of Cno's PDZ and FAB domains, to define how they contribute to Cno localization and function throughout development, thus providing insights into Afadin's roles in mammals. In parallel I will perform protein-protein interaction analysis of Cno mutant proteins in vitro and in vivo. Aim 2 is built on the hypothesis that an active pool of Rap1 regulates Cno activity, and that Dizzy is the predominant Rap1 GEF regulating Cno's localization and thus function during embryogenesis. To test this, I will explore how Rap1, an upstream regulator of Cno, coordinates Cno localization and function. I will compare the phenotype of Cno knockdown with that of Rap1, and with Dizzy and RapGAP1 knockdown, known regulators of Rap1 activity. In parallel I will develop tools revealing where active Rap1 localizes. This study will define how Rap1 activity regulators, Rap1, and Cno work to regulate dynamic cell behaviors and will provide critical information to understand their roles in disease. Through this training, I will gain cutting edge skills in molecular biology, biochemistry, genetics and microscopy, learn to bridge protein structure with domain function, and develop critical presentation and writing skills. These will all further my long-term career goal of becoming a professor and a principal investigator.